COMPARISON · FACTS, NOT SCORES · ACCESSED 2026-08-27

Four honeypot checkers, compared without a thumb on the scale.

A comparison table in which the author wins is the least trustworthy artifact in the category — so this one has no scores, no ranking column and no verdict. Every cell is a documentation-verifiable fact from each tool's own public pages (accessed 2026-08-27); where a tool's pages are silent, the cell says “not documented” — a fact about documentation, not a claim about the tool. Columns are alphabetical.

Disclosure first: the X-Ray's behavioural instrument IS honeypot.is — a third-party public service, named on the tool page itself. Their simulator covers more chains than we expose, and RugCheck owns the Solana ground we do not measure at all. Both are strengths of theirs, stated plainly.

Documented factBlazePhoenix X-Rayhoneypot.isRugCheckTokenSniffer
Behavioural sell simulation (executes a buy AND sell against the real contract)Disclosure: the X-Ray’s behavioural instrument IS honeypot.is — a third-party public service we name on the tool page itself. Their simulator is the upstream; we add the second instrument.Yes — via honeypot.is (disclosed below)Yes — its core functionNot documented as simulation (report-based checks)Static/code analysis; simulation not documented
On-chain routable-liquidity measurement through the tool’s own deployed contractsThe X-Ray’s second, independent instrument: does a sell ROUTE actually exist, measured by the same contract that powers the swap.Yes — previewPlan on the deployed Quoter, reproducible by anyone via eth_callNot documentedNot documentedNot documented
Chains covered (per each tool’s own site)honeypot.is covers more simulation chains than the X-Ray exposes; RugCheck owns the Solana ground the X-Ray does not measure at all. Both stated plainly.Measurement: Base, Ethereum, Optimism, Arbitrum, Robinhood Chain; simulation: Ethereum + BaseEthereum, BSC, Base (broader than our simulation reach — their strength)Solana-focusedMultiple EVM chains
Free web check without API key, signup or wallet connectionYesYesYesYes (advanced features gated)
Method published on the tool page ("what the verdict means and does not mean")Read each tool’s own page and judge; ours is at /xray and /xray/methodology.Yes — instruments named, limits section, HPF taxonomyPartial — simulator describedPartialPartial — audit criteria described
Explicit "what this scanner does NOT do" statementA tool that lists its powers is an advertisement; a tool that lists its limits is an instrument.Yes — five stated limits on the tool pageNot documentedNot documentedNot documented
Effective sell-tax reported from executionYes — via the simulation instrumentYesToken-report fieldsScore components
Embeddable badge that RE-RUNS the check live (vs a static seal)Yes — dated, nofollow, re-verifies on clickNot documentedNot documentedNot documented
Machine-readable access for AI agents (MCP server)Yes — public MCP serverNot documentedNot documentedNot documented
Verdict asymmetry stated (failed sell convicts; empty quote only suspects)Yes — stated on the tool pageNot documentedNot documentedNot documented
Named honeypot-family taxonomy (cite-by-code)Yes — HPF-01…07No equivalent foundNo equivalent foundNo equivalent found
Runs the checks from the reader’s browser (no backend of its own)Neither model is "better" — client-side means nothing is stored by us; server-side can cache and rate-protect. Stated as architecture, not as a score.Yes — both reads client-sideServer-side (their infrastructure)Server-sideServer-side

The method, so you can re-run it

Open each tool's public pages and check every cell — that is the entire methodology, and it works on us too. The one thing no documentation can settle is behaviour, and for that each tool must be run: paste the same token into all four and compare what comes back. The X-Ray's verdict semantics are published at /xray/methodology (the HPF taxonomy), and its stated limits are on the tool page.

Corrections welcome: if any cell misstates a tool's documentation, the fix ships and lands in /corrections.json — this page is subject to the same register as everything else here.

Related

Why the aggregator comparison publishes a METHOD, not a table › · Can't sell a token? The eight causes › · Our own measured record ›